Hi,

I again have some questions about simulation. I would really appreciate if
you could answer them.

1) When I run a benchmark from the splash image (I have not tried PARSEC
yet) on marss.dramsim, the host's terminal shows the increment of
statistical variables such as "cycles" for a while and after that it stops
counting (increasing). This stop occurs few minutes after beginning of the
execution and at that point of the time the benchmark's execution is still
running. For instance, in the case of "contiguous ocean" microbenchmark the
total execution takes about 1 hour while the cycle counting stops in 5-10
mins after the beginning of the execution. So, I wanted to know if this is
something that happens in your simulations as well? Can I trust the
statistics?

2) There is also another issue when I simulate using a command such as
"./start_sim; ls; ./stop_sim". The issue is that after the simulator
receives the "kill" signal when it executes the "stop_sim" and then the
guest machine is turned off. Is this normal? Does this happen for you as
well? If not, please let me know how I can fix the problem.

3) How can I distinguish each microbenchmark's accessed memory addresses
when I run multiple micro-benchmarks simultaneously?

Thanks in advance
_______________________________________________
http://www.marss86.org
Marss86-Devel mailing list
[email protected]
https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel

Reply via email to